NSF 2002 Summer School (Urbana-Champaign)Microfluidics

George Em KarniadakisBrown University

• Basic Concepts and Applications• Gas Flows• Liquid Flows• Particulate Flows• Moving Domains and Applications

Reference: G.E. Karniadakis & A. BeskokMicro Flows: Fundamentals and Simulation,Springer, 2002.

Microfluidics:Emerging technology that allows development of new approaches tosynthesize, purify, and rapidly screen chemicals, biologicals, and materialsusing integrated, massively parallel miniaturized platforms.

•• Microfluidics is Microfluidics is InterdisciplinaryInterdisciplinary::– Micro-Fabrication– Chemistry– Biology– Mechanics– Control Systems– Micro-Scale Physics and Thermal/Fluidic Transport– Numerical Modeling– Material Science– System Integration and Packaging– Validation & Experimentation– Reliability Engineering– …

Microfluidic Applications:Defense Applications:Defense Applications:

• Lab on a chip: µ-TAS

Bio-Medical Applications:Bio-Medical Applications:• Drug Delivery Systems• DNA Analyzers• Human Health Monitoring• Artificial Organs

Environmental Monitoring:Environmental Monitoring:• Water & Air Pollution Sensing• Gas/Liquid Filtration Systems

Microelectronics:Microelectronics:• Thermal Management• Bubble-Jet Printers

Aerospace Industry:Aerospace Industry:• Drag & Stall Control

Microfluidic Devices

Motion Generation:Micro-Motors, Turbines,Steam Engines, Gears, Pistons, Links

Micro-Total-Analysis-Systems (µ -TAS)seamlessly integrate sample collectionand separation units, biological andchemical sensors, fluid pumping andflow control elements, and electronicson a single microchip.

Sensors & Actuators:Pressure, Temperature, Shear Stress,Biological & Chemical Sensors

Fluidic Components:Channels, Pumps, MembranesValves,Nozzles, Diffusers and Mixers

Scientific & Technological ChallengesScientific & Technological Challenges• Development of new concepts that are specifically designed to takeadvantage of the small scale of microfluidic devices,

• To impart unique new functions that are not simply miniaturizedversions of existing systems and components.

Mass Flow Rate versus Pressure DropMass Flow Rate versus Pressure Drop

•Linear Scaling •Quadratic Scaling

•Pipe Flow (2mm x 200 mm; gas at low pressure)

Deviations from Continuum - GasesDeviations from Continuum - Gases

heliumnitrogen

•Microchannel: 0.51 microns (Bau et al., U Penn, 1988)•C*=Poex/Poth where Po=Cf Re•Po=64 (pipe)• Po=96 (2D channel)

Deviations from Continuum - LiquidsDeviations from Continuum - Liquids

•Microducts (Bau & Pfahler, 2001)•Silicone oil

•Question: Anomalous Diffusion or something else?

Interface Inside Carbon Nano-Tubes

Courtesy of Megaridis & Gogotsi, UIC

(transmission electron micrographs)

T-junction* Quake, Caltech in

in out

Applications of Particulate Microflows

Micro fluorescent activated cell sortingµFACS

Micro magnetic activated cell sorting

µMACS

-Sorting-Analyzing-Removal

•Cells•Particles•Embryos

OfLiquids + Gas

* Telleman et al.

Characterization of Airborne Particles

Anthrax Spores Anthrax Bacteria

Size, Shape & Orientation

Active Control of Supraparticle Structures Microchannels

Hayes et al. Langmuir (2001)

Entropic Trapping and Seiving of DNA in Nanofluidic Channels

Han & Craighead, Science, (2000)

Courtesy of Texas Instruments

Digital Micro-Mirror DeviceDigital Micro-Mirror Device TM

Digital Light Processing TM:848 x 600 pixels1280 x 1024 pixels16µm x 16µm w/ 1 µm separation

DMD 101 DMD 101 -- Intro to DMD & DLP Technology (3Intro to DMD & DLP Technology (3--hour version) hour version) -- Section 04 Section 04 -- Architecture & Electromechanical Operation Architecture & Electromechanical Operation (LJH:10/19/00)(LJH:10/19/00)

Slide 4Slide 4--1919DMD Analog Transient ResponseDMD Analog Transient ResponseEffect of “Air” Damping (circa 1988)

840 X 1 DMD19 µm mirrors

Time (Time (µµS)S)

Atm.Atm.

8585

2525

11Pres

sure

(mm

Hg)

Pres

sure

(mm

Hg)

13.6 13.6 µµSS

11.6 11.6 µµSS

AddressAddressPulsePulse

20 20 µµSS

OpticalOpticalResponseResponse

DMDTM Air Damping Effects & Transient ResponseCourtesy of Texas Instruments

Pressure

Simulation of Micro-Pulse-Plasma Thrusters (micro-PPT)

• New issues arise in simulating plasma microthrusters that are an order ofmagnitude smaller than the existing state-of-the art. – Current simulation technologies are based on mathematical models and

assumptions that break down in microscales and, therefore, cannot treatthese microflows comprehensively.

• Microspacercaft are currently considered by Air Force, NASA and industryfor a variety of applications.– L<10 cm, W< 1 kg, and P< 10 W. – Need for micropropulsion. – Plasma microthrusters introduce an additional complexity in their

analysis due to the overlap of electrodynamic and gasdynamic scales.

Physical Processes in a micro-PPT

•Gas DischargeProcesses

•MagnetogasdynamicAcceleration

•Coupling withExternal Circuit

•Teflon Ablation

AFRL micro-PPTs

Gas Microflows: Typical Applications

2Kn

Lλ γπ

= =2 Re

MKnLλ γπ

= =

Knudsen number

• How small should a volume of fluid be sothat we can assign it mean properties?

• At what scales will the statisticalfluctuations be significant?

• Are the low-pressure rarefied gas flowsdynamically similar to the gas micro-flows?

The Continuum Hypothesis

n/n0

L(m

icro

ns)

10-4 10-3 10-2 10-1 100 101 10210-3

10-2

10-1

100

101

102

103

Kn = 1.0

Kn = 10

L / = 20δ

Dilute Gas Dense Gas

Kn = 0.1

Kn=0.01

S lipFlow

Trans itionalFlow

FreeMolecularFlow

ContinuumFlow

Navier-S tokesEquations

The Continuum Hypothesis

Significant fluctuations

(number density)

LIQUIDS:LIQUIDS: Nano Nano-Scale Behavior-Scale Behavior

Layering of Lennard-Jones molecules near a smooth surface.

Density fluctuations across a nano-channel.

MD Simulations (Koplik & Banavar, ARFM 1995)

wall

fluid

•3D periodic channel 51.30x29.7x25.65 (molecular units)•27,000 number of atoms•2,592 atoms of each wall (FCC lattice type)•1 atom = 1 unit

The Continuum Hypothesis

classicaldisordered>>1weakgas

quantum+classical

Semi-orderO(1)MediumLiquid

quantumordered<<1StrongSolid

statisticsStructure(molec)

Motion(d_0)

Force(molec)

Type

•(Batchelor’s book)

Slip Length in Complex Liquids(Thompson & Troian, Nature, 1998)

0 [1 ]s

s c

LL

αγγ

= −

•MD of Couette flow•Lennard-Jones•H=25.57σ

•Slip length increases as wall-energydecreases or wall-density increases•For slip length > 17σ :strong nonlinear response?

Question: At high shear rates, is theliquid behavior near the wall non-Newtonian?

Physical Challenges of Micro-Scale TransportPhysical Challenges of Micro-Scale Transport•• Gas FlowsGas Flows

– Compressibility– Rarefaction

• Slip• Transition• Free Molecular• Thermally Induced Motion

– Surface & Roughness– Viscous Heating– Incomplete Similitude– …

•• Liquid FlowsLiquid Flows– Wetting– Adsorption– Slip– Electrokinetics– Polarity– Coulomb & van der

Waals Forces– Capillary Forces– Roughness– …

Constitutive Laws,Constitutive Laws,Boundary Conditions,Boundary Conditions,Surface, Interface andSurface, Interface andBody ForcesBody Forces

Micro Flows: Fundamentals & SimulationMicro Flows: Fundamentals & SimulationKarniadakis & Beskok, Springer, New York, 2001ISBN 0-387-95324-8

Numerical Modeling ChallengesNumerical Modeling Challenges

• Multi Physical Phenomenon (Thermal, Fluidic, Mechanical, Biological, Chemical, Electrical)

• Multi Scale (Atomistic, Continuum)• Complex Geometry

Numerical Simulation StrategiesNumerical Simulation Strategies

4 Scientific Simulations:• Multi Scale• Simpler Geometry• Accurate (error < 1 %)

4Engineering Simulations:• Multi Scale• Multi Physics• Complex Geometry• Accurate (error < 5~10 %)

4Low Order Models:• Multi Physics• Full Device Simulation• Lower Accuracy, but Fast (error ~10 %)

Challenges of Microscale ResearchChallenges of Microscale Research•• Small Scale PhysicsSmall Scale Physics

–– Governing Equations BreakdownGoverning Equations Breakdown• Constitutive Laws• Boundary Conditions

–– Microscopic EffectsMicroscopic Effects• Dominance of Surface Forces• Coulomb Forces• Van der Waals Forces• Capillary Forces• Importance of Molecular Structure

•• Detection & Experimental VerificationDetection & Experimental Verification• Micro Particle Image Velocimetry• Molecular Fluorescence Velocimetry

Numerical Modeling Methods

Experimental LimitationsExperimental Limitations

•Micro-Particle-Image-Velocimetry (Meinhart et al.)

•30 x 300 microns channel•Resolution: 450 nm in the wall-normal

•Overlapped windows•Specialied interrogation algorithms

Prototype Flows• Pressure-Driven Flows

– Poieseuille flow• Shear-Driven Flows

– Couette flow– Cavity flow

• Squeezed Film – Lubrication– Reynolds equation

• Electrokinetically-Driven Flows– Dielectrophoresis

• Thermal Creep– Knudsen compressors

• Surface-Tension-Driven Flows- routing of droplets

•Microfluidics:The path and link to nanotechnology

•The wet circuit liquid circuit + element

•Fluidic Self Assembly – Massively Parallel VLSI-like, programmable fluidic networks with logic

•Fluids with more than Viscosity – MOEMS

•Complex Fluids •Magnetorheological fluids

•Magnetic (high-frequency) microfluidics •Dynamic reconfiguration, use E to change properties

•Wall-less fluidics/interfaces (Bebe, Whitesides)

•Smart dust – Biomimetic sensors

FUTURE

Sandia nanospheres